English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Accurate polyp segmentation is of great importance for colorectal cancerdiagnosis and treatment. However, due to the high cost of producing accuratemask annotations, existing polyp segmentation methods suffer from severe datashortage and impaired model generalization. Reversely, coarse polyp boundingbox annotations are more accessible. Thus, in this paper, we propose a boostedBoxPolyp model to make full use of both accurate mask and extra coarse boxannotations. In practice, box annotations are applied to alleviate theover-fitting issue of previous polyp segmentation models, which generatefine-grained polyp area through the iterative boosted segmentation model. Toachieve this goal, a fusion filter sampling (FFS) module is firstly proposed togenerate pixel-wise pseudo labels from box annotations with less noise, leadingto significant performance improvements. Besides, considering the appearanceconsistency of the same polyp, an image consistency (IC) loss is designed. SuchIC loss explicitly narrows the distance between features extracted by twodifferent networks, which improves the robustness of the model. Note that ourBoxPolyp is a plug-and-play model, which can be merged into any appealingbackbone. Quantitative and qualitative experimental results on five challengingbenchmarks confirm that our proposed model outperforms previousstate-of-the-art methods by a large margin.

BoxPolyp:Boost Generalized Polyp Segmentation Using Extra Coarse Bounding Box Annotations